The current proposal seeks support for research to document and reduce the expression of implicit verbal and nonverbal bias by medical residents when they interact with Hispanic patients. Research shows that while stereotyping, prejudice and acts of discrimination can be conscious and deliberate, the expression of bias among healthcare providers is often unintentional because many hold negative attitudes and stereotypes at an implicit or unconscious level. Thus, implicit verbal and nonverbal biases may leak into the way residents interact with Hispanic patients, but training residents to control thei verbal and nonverbal implicit biases can mitigate the negative effect on patients during a clinical encounter. Whereas previous research shows that healthcare providers hold implicit negative attitudes and stereotypes about stigmatized patients, and that these biases can negatively impact both the way providers interact with minority patients, and the satisfaction that minority patients report after a clinical encounter, no research to date has documented the mediators of the relationship between implicit bias and patient satisfaction. The current proposal addresses this void by developing and testing a new model of Implicit Linguistic Bias. The Implicit Linguisti Bias model proposes that interactions between Hispanic patients and residents, who hold implicit biases toward them and their group are characterized by lower linguistic matching, higher use of demand-withdrawal language, and higher intergroup linguistic bias language. The presence of these implicit linguistic biases, in turn, make the patient feel less involved in the conversation, feel less confidence and trust in the provider, and feel less desire to see the provider again. Reducing the expression of implicit verbal and nonverbal biases, including implicit linguistic biases, during a clinical encounter, however, can be achieved by training residents in the psychology of implicit bias, how it is communicated during a clinical encounter, and how they can control their verbal and nonverbal expressions of implicit bias when they interact with Hispanic patients. In three large-scale studies over five years, the proposed research will examine the presence and reduction of verbal and nonverbal forms of implicit bias, including the effects of implicit linguistic bias, during clinical encounters between medical residents and Hispanic patients. Study 1 will develop and test the proposed model of Implicit Linguistic Bias by recording and coding interactions between medical residents and Hispanic patients. Study 2 will develop and test the effectiveness of a new hybrid-learning module for training medical residents in the psychology of implicit bias. Study 3 will then examine if completing the implicit bias training module reduces the expression of implicit verbal and nonverbal bias, including the expression of implicit linguistic bias, and improves patient satisfaction ratings when medial residents interact with Hispanic patients.